Sign inSign up

vllm/vime:r015mtp-arm64

Index digest

sha256:990f8433bc3481760de18ae8d3987af27bcf86ca36903b39d69b3769ad9de6a0

OS/ARCH

linux/arm64

Compressed size

15.27 GB

Last pushed

3 months by vllm

Type

Image

Manifest digest

sha256:6a555cc87532a714b1693830cfbddd79a8876c426956694bc7825e16bb44c764

Image Layers

Image Layers

1ARG RELEASE0 B
2ARG LAUNCHPAD_BUILD_ARCH0 B
3LABEL org.opencontainers.image.ref.name=ubuntu0 B
4LABEL org.opencontainers.image.version=24.040 B
5ADD file ... in / 28.55 MB
6CMD ["/bin/bash"]0 B
7ENV NVARCH=sbsa0 B
8ENV NVIDIA_REQUIRE_CUDA=cuda>=13.0 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<5360 B
9ENV NV_CUDA_CUDART_VERSION=13.0.96-10 B
10ARG TARGETARCH0 B
11LABEL maintainer=NVIDIA CORPORATION <[email protected]>0 B
12RUN |1 TARGETARCH=arm64 /bin/sh -c4.32 MB
13ENV CUDA_VERSION=13.0.20 B
14RUN |1 TARGETARCH=arm64 /bin/sh -c95.11 MB
15RUN |1 TARGETARCH=arm64 /bin/sh -c186 B
16ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin0 B
17ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
18COPY NGC-DL-CONTAINER-LICENSE / # buildkit6.72 KB
19ENV NVIDIA_VISIBLE_DEVICES=all0 B
20ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility0 B
21ARG CUDA_VERSION0 B
22ARG PYTHON_VERSION0 B
23ARG DEADSNAKES_MIRROR_URL0 B
24ARG DEADSNAKES_GPGKEY_URL0 B
25ARG GET_PIP_URL0 B
26ENV DEBIAN_FRONTEND=noninteractive0 B
27WORKDIR /vllm-workspace104 B
28RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=230 B
29RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=215.94 MB
30RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=1.81 GB
31RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=48.51 MB
32ENV UV_HTTP_TIMEOUT=5000 B
33ENV UV_INDEX_STRATEGY=unsafe-best-match0 B
34ENV UV_LINK_MODE=copy0 B
35ENV UV_PYTHON_INSTALL_DIR=/opt/uv/python0 B
36ENV UV_CACHE_DIR=/opt/uv/cache0 B
37RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=150 B
38RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=3.62 KB
39COPY docker/entrypoints/vllm-nonroot-entrypoint.sh /usr/local/bin/vllm-nonroot-entrypoint.sh # buildkit2.03 KB
40RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=2.03 KB
41ENV VLLM_ENABLE_CUDA_COMPATIBILITY=00 B
42ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
43COPY requirements/common.txt /tmp/common.txt # buildkit1.64 KB
44COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit648 B
45RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=3.66 GB
46ARG FLASHINFER_VERSION=0.6.120 B
47RUN |7 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=1.63 GB
48ARG GDRCOPY_CUDA_VERSION=12.80 B
49ARG GDRCOPY_OS_VERSION=Ubuntu22_040 B
50ARG TARGETPLATFORM0 B
51COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit896 B
52RUN |10 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=519 KB
53ARG BITSANDBYTES_VERSION_X86=0.46.10 B
54ARG BITSANDBYTES_VERSION_ARM64=0.42.00 B
55ARG TIMM_VERSION=>=1.0.170 B
56ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.70 B
57RUN |14 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=300.17 MB
58ARG PIP_INDEX_URL UV_INDEX_URL0 B
59ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
60ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
61ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER0 B
62ARG PYTORCH_NIGHTLY0 B
63COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit198 B
64COPY /workspace/dist/wheel.sha256 /tmp/vllm-wheel.sha256 # buildkit225 B
65RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=464.33 MB
66RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=32 B
67ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
68COPY /tmp/ep_kernels_workspace/dist/wheels.sha256 /tmp/ep-kernels-wheels.sha256 # buildkit253 B
69RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=10.8 MB
70RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=78.79 MB
71RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=429.92 MB
72ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
73COPY examples examples # buildkit257.86 KB
74COPY benchmarks benchmarks # buildkit249.61 KB
75COPY ./vllm/collect_env.py . # buildkit9.62 KB
76ARG TARGETPLATFORM0 B
77ARG INSTALL_KV_CONNECTORS=false0 B
78ARG CUDA_VERSION0 B
79ARG VLLM_BUILD_COMMIT0 B
80ARG VLLM_BUILD_PIPELINE0 B
81ARG VLLM_BUILD_URL0 B
82ARG VLLM_IMAGE_TAG0 B
83ARG PIP_INDEX_URL UV_INDEX_URL0 B
84ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
85ENV UV_HTTP_TIMEOUT=5000 B
86ARG torch_cuda_arch_list=7.5 8.0 8.6 8.90 B
87ENV TORCH_CUDA_ARCH_LIST=8.0 8.7 8.9 9.00 B
88RUN |12 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=13.0.2774.36 MB
89ARG MOONCAKE_WHEEL_AARCH640 B
90ARG MOONCAKE_WHEEL_X86_640 B
91RUN |14 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=13.0.2425.35 MB
92ENV VLLM_USAGE_SOURCE=production-docker-image0 B
93ENV VLLM_BUILD_COMMIT=ee0da84ab9e04ac7610e28580af62c365e898389 VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/3274 VLLM_IMAGE_TAG=vllm/vllm-openai:v0.24.0-ubuntu24040 B
94LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=ee0da84ab9e04ac7610e28580af62c365e898389 org.opencontainers.image.version=vllm/vllm-openai:v0.24.0-ubuntu2404 org.opencontainers.image.url=https://buildkite.com/vllm/release-v2/builds/32740 B
95ENTRYPOINT ["vllm" "serve"]0 B
96ARG PATCH_VERSION=latest0 B
97ARG MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e0870 B
98ARG ENABLE_CUDA_13=10 B
99WORKDIR /root/32 B
100RUN |3 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=11.4 GB
101RUN |3 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1177 B
102RUN |3 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1705.54 MB
103RUN |3 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=11.41 GB
104RUN |3 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=11.6 MB
105RUN |3 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=12.47 MB
106ARG INSTALL_FLASHQLA=00 B
107RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=132 B
108RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=132 B
109RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1605.24 MB
110RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1250.74 MB
111RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=147.75 MB
112RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1118.87 MB
113RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1366.6 KB
114RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=12.64 MB
115RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1111.94 MB
116COPY requirements.txt /tmp/requirements.txt # buildkit398 B
117RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1488.02 MB
118RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=132 B
119RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1101.89 MB
120RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1176 B
121COPY docker/patch/latest/megatron.patch /root/Megatron-LM/ # buildkit11.16 KB
122RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=1109.45 MB
123COPY docker/patch/latest/vllm.patch /tmp/vllm.patch # buildkit1.2 KB
124RUN |4 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=113.94 MB
125ARG VIME_COMMIT=main0 B
126RUN |5 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=145.95 MB
127RUN |5 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 ENABLE_CUDA_13=11.43 MB
128ENTRYPOINT []0 B
129CMD ["/bin/bash"]0 B
130COPY api_router.py /tmp/api_router.py # buildkit2.04 KB
131RUN /bin/sh -c VLLM_SITE="$(python3 -c2.31 KB
132COPY overlay_46725.patch /tmp/overlay_46725.patch # buildkit2.83 KB
133RUN /bin/sh -c VLLM_SITE="$(python3 -c8.12 MB

Command

 ARG RELEASE